Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Crm Agents
v1.0.0Execute 13 CRM AI agents on demand — oracle, spider, watchdog, diplomat, strategist and more. Business intelligence as a service.
⭐ 0· 96·0 current·0 all-time
byOrion@drivenbymyai-max
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The name/description (CRM AI agents) matches the SKILL.md content: it documents endpoints to list and run agents and describes agent types. Calling an external service to run agents is coherent with the stated purpose. However, the skill mentions paid execution and a homepage (sputnikx.xyz) in the SKILL.md metadata while the registry metadata earlier listed no homepage/source — this metadata inconsistency reduces trust.
Instruction Scope
All runtime instructions direct the agent to POST tasks and queries to https://sputnikx.xyz/api/v1/agent/* via curl. That is expected for an external SaaS-backed skill, but the SKILL.md does not specify how to authenticate, how payment is authorized, or what data is required/allowed. Because the agent will forward whatever task/context is provided to a third-party endpoint, there's a real risk of unintentionally sending sensitive or private data to an external service.
Install Mechanism
This is an instruction-only skill with no install spec and no code files — nothing is written to disk and no third-party packages are fetched. That minimizes local install risk.
Credentials
The SKILL.md declares no required environment variables or credentials, yet the examples show paid execution (USDC) and calls to a third-party API. The absence of declared auth tokens/keys or instructions for payment authorization is disproportionate/unexplained for a networked paid service and reduces transparency about what secrets or accounts the service will require when actually used.
Persistence & Privilege
The skill is not forced always-on and does not request special system persistence. Agent autonomous invocation is allowed by default but not by itself a red flag here. There is no indication the skill modifies other skills or system settings.
Scan Findings in Context
[NO_CODE_FILES] expected: The regex scanner found no code to analyze because this is an instruction-only skill (SKILL.md contains curl examples). Network calls are present in the instructions rather than programmatic code.
What to consider before installing
This skill will send whatever task/context you give it to an external domain (https://sputnikx.xyz) and the instructions mention a payment cost but do not explain authentication or how payments are handled. Before installing or using it: 1) Verify the service owner and source code or privacy/security documentation for sputnikx.xyz; 2) Do not send any sensitive data (PII, credentials, customer records) in task text until you confirm encryption, retention, and billing terms; 3) Ask the publisher how payments are authorized (wallet signature? API key?) and whether any secrets will be required; 4) Test first with non-sensitive sample queries; and 5) If you manage a corporate agent, consider network egress controls or policy approval before allowing calls to this third-party API.Like a lobster shell, security has layers — review code before you run it.
latestvk979w4syzrqxnw4etw0tme8f5h83dp57
License
MIT-0
Free to use, modify, and redistribute. No attribution required.